Tactile Controls In A Digital World

This article was written by Scott Jenson and Michael DiTullo and published at Core77 in April 2024

A few recent tech writers have leaked that the new AirPods case will likely have a touch screen. Other earbud makers have tried this as well but it’s Apple, so people will naturally have strong opinions, and we’re no different. Designers always see what could be, and the two of us having worked with brands like Apple, Google, Motorola, and Nike had some thoughts on what it takes to make a beautiful and intuitive tactile experience in an increasingly digital world. While Apple may or may not decide to put a screen on the AirPod case, it’s unfortunately the expected, even safe answer for almost any electronic product today. But is it the right choice? That’s the purpose of this exploration, to push at this assumption and see if we can find something more useful and even delightful.
There are clearly advantages to using a touchscreen:

1) It’s more flexible: many screens and many functions.
2) It can be updated: anything can be changed later.

These are big advantages but they are mostly practical, covering a broad range of features. There’s nothing wrong with that but we want to explore something more visceral and urgent. Instead of many screens and functions, let’s focus on what users actually need frequently. In addition, let’s go beyond a generic “swipe and tap” interaction style to something more nuanced and analog. 
Moving from a digital screen to analog controls is an interesting challenge. It may be a bit more expensive and likely a bit harder to build but those are secondary concerns that can be solved later. Let’s not limit our vision too quickly. Instead of being driven by constraints, what if we leaned into analog controls to see if we can find something far more interesting? 

First let’s point out the elephant in the room: an analog approach will be more limited functionally. That’s why most screens exist today, whether it’s on your smart fridge or your EV car: screens are popular as you can cram dozens of small buttons on them. But as we said before, this is mostly a production concern, it ignores the many ergonomic, safety, and even sensual aspects of physical controls. But don’t take our word for it, there is already a backlash forming against digital screens in cars today

Early sketch iterations on what a physical layout might be like

Given how common it is to think in terms of long feature lists, how can we justify a more limited approach? Isn’t missing any amount of functionality a liability? It depends on your point of view: Engineers tend to think mathematically while Designers tend to think statistically. Both are correct, they just have different goals. If your ONLY goal is to cover every possible feature, then having a touchscreen is reasonable. However, if you think statistically and ask “What do people do 99% of the time?” you get a much different answer. 

That’s where we started: what do people need to do the vast majority of the time?
 * Adjust the volume
 * Adjust tracks forward and backward
 * Play/Pause (and depending on context answer/hang up the phone)
The default approach is to have three simple push buttons: 

Classic physical 3 button transport controls Apple pioneered on their earpods which became industry standard

Nothing is inherently wrong, it’s quite minimalist. But in order to squeeze in extra functionality buttons are “doubled up”, so, for example, double tapping the plus button would go to the next track. This isn’t just hidden, it’s error prone. It may work “on paper” but it’s tedious and what we’ve all had to endure with cheap digital audio devices for years. We can do better.

Our approach is to get rid of this ‘doubling up’ by adding two additional buttons so each has just a single function, making things easier to learn, reducing errors and allowing each button to physically express what it does. This allows us to create buttons with more character: labels aren’t always enough, it’s helpful if buttons actually look the part. We feel that a device should physically tell you how to use it, whether reading the labels or by touch when it is in your jacket pocket.

Michael and Scott’s explorations, from the left to right: Current design, standard 3 button, Multi Function Button (MFB) with single rocker switch, volume roller with USB and single rocker, volume roller with MFB and rocker

In addition, we wanted to get rid of the ‘discrete interaction model’ for volume. Instead of tapping the + button 6 times, overshooting and then tapping the – button twice, it’s far easier, faster, and pleasurable to use a roller. There is a reason high-end stereos have buttery smooth dials instead of clicky +/- buttons.
To be fair, the play/pause button is slightly multi-function as it also answers/ends phone calls but that is driven by a clear context: if a call comes in, the button answers the call, if not, it plays music. What is harder to ignore is activating the voice assistant. Here we chose to make a compromise and double up the play/pause button: have long-press start the voice assistant. This actually mirrors what the earbuds do today so felt compatible with existing user expectations. Our goal was to avoid long-press functions but this feels like a reasonable compromise.
Using these assumptions, this is our proposed design:

Scott and Michael’s final design

Note the layout is not symmetric. The goal is to be able to hold this inside your pocket without looking at it and know exactly what button to press. Each button is indented into the case to prevent accidental use (this would need to be tested) Also notice that we’ve added an LED into the Play/Pause button to replace the existing “Power LED” on Airpods today. This is a slight nod to manufacturing simplicity (one less case integration). At the very least, it’s the same LED on the case today to indicate charging status. But it’s also an opportunity to explore other types of feedback. For example, if the LED was multi-colored, it could pulse red if the music was paused.
This transforms the case from a utilitarian shell into a fidget-like device that you’d want to hold in your hand. 

Scott and Michael’s final design in use

We may be doing our best to create a digital world as quickly as possible, but that won’t replace hundreds of thousands of years of evolution making us deeply physical beings. From the time we are infants in our crib we humans love to touch stuff. We love tactility, shape, texture, and color. As designers we don’t want to fight that, we want to work with it! We think the world can be better, or at the very least more friendly and enjoyable. So to our fellow hardware designers, developers, engineers, and product managers out there who might be reading this, we dare you to take the road less traveled and make something that people will truly love to use.

About the authors:
Scott Jenson:
Scott Jenson started as a UX designer at Apple in 1988, working on System 7, Newton, and the Apple Human Interface guidelines. He moved to London to be Director of Symbian’s DesignLab, then joined Google in 2005 where he designed the first version of Google Mobile Maps. He went on to manage the Mobile UX team at Google, then left to be a creative director at frog design in San Francisco. He headed up design for two startups and eventually returned to Google to work in Chrome and Android. He has over 35 patents and is now semi-retired.

Michael DiTullo:
Michael DiTullo has been designing iconic products for some of the world’s best brands for more than 25 years. He has worked with an amazingly wide assortment of companies including Nike, Google, Honda, Timex, Chantal, Converse, Motorola, Hasbro, Arc Electric Boats and Kirei. Prior to starting his eponymous design studio Michael was Chief Design Officer for Sound United, creative director for frog design’s San Francisco studio and spent nearly a decade at Nike. Michael is listed on over 30 patents and has won numerous awards including the IDSA’s special lifetime achievement award for contributions to the design industry.